是否按组显示月度趋势？

Question

我有一个数据框，看起来像这样：

df <- read.table(text = "ID    Date      Freq      Gender      Trt      month   Group
    15 A1 2019-01-29     1       female       1          1      G1                                  
    16 A1 2019-04-27     10      female       1          4      G1                                  
    17 A1 2019-04-27     10      female       1          4      G1                                     
    18 A1 2019-04-29     2       female       1          4      G1                                
    19 A1 2019-04-29     12      female       1          4      G1                                      
    20 A1 2019-10-09     6       female       1          10     G1                                      
    21 A1 2019-12-26     NA      female       1          12     G1                                     
    22 A1 2019-12-26     NA      female       1          12     G1                                     
    27 A1 2019-02-19     5       female       1          2      G1                                    
    28 A2 2019-02-19     15      male         0          2      G3
    37 A3 2019-01-30     NA      female       0          1      G2
    38 A3 2019-01-30     9       female       0          1      G2
    39 A3 2019-02-15     11      female       0          2      G2
    40 A3 2019-02-15     10      female       0          2      G2
    41 A3 2019-03-16     1       female       0          3      G2
    42 A3 2019-04-01     1       female       0          4      G2
    43 A3 2019-04-01     3       female       0          4      G2
    44 A3 2019-04-22     2       female       0          4      G2
    45 A3 2019-05-25     4       female       0          5      G2
    46 A3 2019-05-29     4       female       0          5      G2
    47 A3 2019-06-06     19      female       0          6      G2")

我想逐月展示不同组别G1-6和trt的趋势。例如，xbar上有1-12个月的图表，ybar上有按G分组的该月人数的计数(因此，在图表上为G1、G2等绘制不同颜色的趋势线，显示该月的组数是如何下降或上升的)，或者甚至是另一种形式的描述性统计数据将会很有帮助。但我想不出如何让任何图形工作。我已经为年份制作了单独的专栏，并尝试为组制作专栏，但它就是不能绘制图表？以下是我所做的和尝试的：

df$Date <- as.Date(df$Date)
dfDates <- data.frame(year = as.numeric(format(df$Date, format = "%Y")), month = as.numeric(format(df$Date, format = "%m")))
dfDates <- cbind(df,dfDates)

dfG <- df %>%
  pivot_longer(cols = c("G1", "G2", "G3", "G4","G5","G6"), names_to = "Group", names_ptypes = list(Group = factor()), values_to = "Value") %>%
  filter(Value == 1) %>%
  select(-Value)

ggplot(dfG, aes(factor(month), sum(Group), group = Group, Trt)) +
  geom_line(size = 1, alpha = 0.8) +
  geom_point(size = 2) +
  scale_color_brewer(name = "Name", palette = "Set1")+xlab("month") + ylab("trend")

ggplot(dfG, aes(factor(month), Group)) +
  geom_line(size = 1, alpha = 0.8) +
  geom_point(size = 2) +
  scale_color_brewer(name = "name", palette = "Set1")+xlab("month") + ylab("trend")

我非常感谢你的时间和帮助，谢谢你。

Answer 1

最简单的方法是首先创建摘要，然后在管道中或首先在您的环境中定义一个新对象将其传递给ggplot()。(我更喜欢)

library(tidyverse)

df2 <- 
  df %>%
  mutate(month = factor(month.abb[month], levels = month.abb)) %>%
  group_by(month, Group) %>%
  summarise(count = max(row_number())) 
#> `summarise()` has grouped output by 'month'. You can override using the `.groups` argument.

ggplot(df2, aes(month, count)) +
  geom_line(aes(color = Group, group = Group), size = 1, alpha = 0.8) 

​

​